Salesloft Prod Checklist

Rate Limiting

• Cost-based budget calculated for expected volume
• Deep pagination avoided (page > 100 costs 3-30x)
• Bulk operations use p-queue or similar throttle
• Rate limit headers logged for capacity planning

Monitoring & Alerting

// Health check endpoint
app.get('/health', async (req, res) => {
  try {
    const start = Date.now();
    await api.get('/me.json');
    res.json({
      status: 'healthy',
      salesloft: { connected: true, latencyMs: Date.now() - start },
    });
  } catch {
    res.status(503).json({ status: 'degraded', salesloft: { connected: false } });
  }
});

• Health check includes SalesLoft connectivity
• Alert on 5xx error rate > 5/min (P1)
• Alert on 429 rate > 10/min (P2)
• Alert on auth failure (P1 -- token may be revoked)
• Latency p99 tracked (baseline: 300ms reads, 500ms writes)

Data Integrity

• Idempotency keys on all create/update operations
• Duplicate detection by email before person creation
• Webhook events deduplicated by event ID
• Audit log captures all API mutations

Rollback Procedure

# 1. Revert deployment
kubectl rollout undo deployment/salesloft-integration
# or: git revert HEAD && git push

# 2. Verify old version healthy
curl -f https://app.example.com/health

# 3. Pause any running cadence syncs
# 4. Notify sales team of rollback

Post-Launch Verification

# Smoke test production endpoints
curl -s -H "Authorization: Bearer $PROD_TOKEN" \
  https://api.salesloft.com/v2/me.json | jq '.data.email'

curl -s -H "Authorization: Bearer $PROD_TOKEN" \
  'https://api.salesloft.com/v2/people.json?per_page=1' | jq '.metadata.paging.total_count'

Error Handling

Resources

• SalesLoft Status
• API Logs

Next Steps

For version upgrades, see salesloft-upgrade-migration.